> -----Original Message-----
> From: Vemula, Hari KumarX
> Sent: Monday, January 7, 2019 1:01 PM
> 
> 
> Signed-off-by: Hari Kumar Vemula <hari.kumarx.vem...@intel.com>
> ---
>  drivers/net/bonding/rte_eth_bond_pmd.c | 11 +++++++----
>  1 file changed, 7 insertions(+), 4 deletions(-)
> 
> -     memcpy(&dev_info->rx_desc_lim, &internals->rx_desc_lim,
> -            sizeof(dev_info->rx_desc_lim));
> -     memcpy(&dev_info->tx_desc_lim, &internals->tx_desc_lim,
> -            sizeof(dev_info->tx_desc_lim));
> +     dev_info->rx_desc_lim = bond_lim;
> +     dev_info->tx_desc_lim = bond_lim;


I relooked into this, these values should be filled from salve data like the 
way done for max_nb_rx_queues and max_nb_tx_queues
Existing code snippet:
if (slave_info.max_rx_queues < max_nb_rx_queues)
                          max_nb_rx_queues = slave_info.max_rx_queues;

 if (slave_info.max_tx_queues < max_nb_tx_queues)
                        max_nb_tx_queues = slave_info.max_tx_queues;

So, something like below you should add for rx/tx desc limit I guess.

if (slave_info.rx_desc_lim.nb_max < max_rx_desc_lim)
                                max_rx_desc_lim = slave_info.rx_desc_lim.nb_max;

if (slave_info.tx_desc_lim.nb_max < max_tx_desc_lim)
                                max_tx_desc_lim = slave_info.tx_desc_lim.nb_max;

dev_info->rx_desc_lim.nb_max = max_rx_desc_lim;
        dev_info->tx_desc_lim.nb_max = max_tx_desc_lim;

@Williams/Declan: Does this make sense?

Thanks,
Reshma

Reply via email to